The following 12 data visualizations show weekly bus ridership in three of Connecticut's largest cities and bus systems for the weeks of 3/01-3/07 to 7/12-7/18.
Because of the large range in ridership numbers between high-volume and low-volume stops in each of the bus systems, we have separated the ridership into chunks using natural breaks found through the Fisher-Jenks optimization method. This data-clustering method is designed to determine the best arrangement of values into different classes. This is done by seeking to minimize each class's average deviation from the class mean, while maximizing each class's deviation from the means of the other groups.
Hartford: 0.0, 707.0, 3654.0, 7295.0, 14397.0
New Haven: 0.0, 272.0, 1263.0, 4137.0, 9711.0
Stamford: 0.0, 246.0, 812.0, 1984.0, 3996.0